157082 research outputs found
Sort by
Comparative Evaluation of Reinforcement Learning and Model Predictive Control for 6DoF Position Control of an Autonomous Underwater Vehicle
No full textAutonomous Underwater Vehicles (AUVs) require precise and robust control strategies for 3D pose regulation in dynamic underwater environments. In this study, we present a comparative evaluation of model-free and model-based control methods for AUV position control. Specifically, we analyze the performance of neural network controllers trained by three Reinforcement Learning (RL) algorithms---Proximal Policy Optimization (PPO), Twin Delayed Deep Deterministic Policy Gradient (TD3), and Soft Actor-Critic (SAC)---alongside a Model Predictive Control (MPC) baseline. We train our RL methods in a simplified AUV simulator implemented in PyTorch, while our evaluation is done in a realistic marine robotics simulator called Stonefish. Controllers are evaluated on the basis of tracking accuracy, robustness to disturbances, and generalization capabilities. Our results show that, MPC suffers from unmodeled dynamics such as disturbances, whereas RL demonstrates adaptation capabilities to disturbances. Also, although MPC demonstrates strong control performance, it requires an accurate model, high compute power and a careful implementation to run in real-time whereas the control frequency of RL policies is only bound by the inference time of the policy network. Among RL-based controllers, PPO achieves the best overall performance, both in terms of training stability and control accuracy. This study provides insight into the feasibility of RL-based controllers for AUV position control, offering guidance for selecting suitable control strategies in real-world marine robotics applications
Effects of semi-rigid behavior on the seismic performance of the precast shear wall with a box-shaped connection joint
No full textBolted connections with obvious semi-rigid characteristics will experience additional moments caused by the second-order effects in structures, leading to inferior seismic performance of precast shear walls compared to cast-in-place shear walls. In order to investigate the influence of semi-rigidity, a new box-shaped connection with apparent semi-rigid features has been proposed. Cyclic loading tests were conducted on a full-scale shear wall specimen (BPSW) with a box-shaped connection and a cast-in-place shear wall specimen (SW1). Failure characteristics, seismic performance, and deformation composition were comprehensively analyzed. Test results indicate that that both BPSW and SW1 exhibit the same bending failure mode. The seismic performance, as indicated by the hysteretic envelope area in BPSW, is slightly smaller than that of SW1. The reduced ductility, initial stiffness, and energy dissipation in BPSW can be attributed to the semi-rigid behavior, which affects the deformation behavior of the precast shear wall structure. Additionally, the deformation composition analysis reveals that rotational deformation of the new box-shaped connection in BPSW cannot be ignored as it accounts for more than 30% of the horizontal displacement. In accordance with previous models, a theoretical model for rotational deformation that offers insights into the semi-rigid nature and seismic performance of the box-shaped connection is proposed. The proposed model demonstrates good agreement with experimental results and provides valuable insights into the semi-rigid behavior and seismic response of such precast connections.</p
A Comparative Analysis of Transformer Models in Social Bot Detection
No full textSocial media has become a key medium of communication in today's society. This realisation has led to many parties employing artificial users (or bots) to mislead others into believing untruths or acting in a beneficial manner to such parties. Sophisticated text generation tools, such as large language models, have further exacerbated this issue. This paper aims to compare the effectiveness of bot detection models based on encoder and decoder transformers. Pipelines are developed to evaluate the performance of these classifiers, revealing that encoder-based classifiers demonstrate greater accuracy and robustness. However, decoder-based models showed greater adaptability through task-specific alignment, suggesting more potential for generalisation across different use cases in addition to superior observa. These findings contribute to the ongoing effort to prevent digital environments being manipulated while protecting the integrity of online discussion
A data augmentation strategy for deep neural networks with application to epidemic modelling
No full textIn this work, we integrate the predictive capabilities of compartmental disease dynamics models with machine learning’s ability to analyze complex, high-dimensional data and uncover patterns that conventional models may overlook. Specifically, we present a proof of concept demonstrating the application of data-driven methods and deep neural networks to a recently introduced Susceptible-Infected-Recovered type model with social features, including a saturated incidence rate, to improve epidemic prediction and forecasting. Our results show that a robust data augmentation strategy trough suitable data-driven models can improve the reliability of Feed-Forward Neural Networks and Nonlinear Autoregressive Networks, providing a complementary strategy to Physics-Informed Neural Networks, particularly in settings where data augmentation from mechanistic models can enhance learning. This approach enhances the ability to handle nonlinear dynamics and offers scalable, data-driven solutions for epidemic forecasting, prioritizing predictive accuracy over the constraints of physics-based models. Numerical simulations of the lockdown and post-lockdown phase of the COVID-19 epidemic in Italy and Spain validate our methodology.</p
Seismic Performance and Mechanical Behavior Assessment of Demountable Diagonal Connection RCS Joints:A Numerical Simulation Study
No full textTo evaluate the performance of the newly designed demountable reinforced concrete column-steel beam (RCS) joint, numerical simulations were performed using the finite element software ABAQUS. The analysis results show that the proposed demountable RCS joint offers enhanced load-bearing capacity and improved ductility relative to conventional cast-in-place joints. A parametric analysis was performed to further investigate the seismic behavior of these joints, focusing on factors such as axial compression ratio, steel beam web strength, stirrup ratio, flange thickness, and Y-shaped connecting plate thickness. Additionally, an analysis of the seismic force transfer mechanism of the proposed joints was conducted. The existing shear capacity calculation formula for RCS joints was improved by considering the components within the joint domain. The improved formula demonstrated a more accurate assessment of the shear capacity of the novel joints, providing a theoretical foundation for future research on this type of joint.</p
Interference Mitigation in Multibeam Satellite Networks as an Optimal Sublattice Problem
No full textResource distribution in radio networks aims at maximizing spectrum utilization while minimizing interference. In this paper, we consider the problem of uniform radio resource distribution on a periodic grid. We formulate the problem as finding the sublattice configuration that maximises the distance between adjacent resources, crucial for reducing interference and improving throughput performance. Leveraging concepts from lattice theory and discrete geometry, we present an enumerative, parallelizable algorithm to explore all possible sublattices and efficiently identify the optimal configurations. Additionally, we investigate the existence and properties of scaled-rotated sublattices, exploring how different lattice geometries impact optimal solutions. Numerical results demonstrate the effectiveness of the proposed algorithm and highlight insights into optimal sublattice design for various lattice structures. Furthermore, the results are applied to the identification of the beam layout in a fixed multibeam geostationary satellite. Numerical results show that the spectral efficiency of the optimised sublattice is higher than all other sublattices. This work thus advances the field of radio resource distribution and offers practical implications for improving satellite network performance
BREA-Depth: Bronchoscopy Realistic Airway-Geometric Depth Estimation
No full textMonocular depth estimation in bronchoscopy can significantly improve real-time navigation accuracy and enhance the safety of interventions in complex, branching airways. Recent advances in depth foundation models have shown promise for endoscopic scenarios, yet these models often lack anatomical awareness in bronchoscopy, overfitting to local textures rather than capturing the global airway structure–particularly under ambiguous depth cues and poor lighting. To address this, we propose Brea-Depth, a novel framework that integrates airway-specific geometric priors into foundation model adaptation for bronchoscopic depth estimation. Our method introduces a depth-aware CycleGAN, refining the translation between real bronchoscopic images and airway geometries from anatomical data, effectively bridging the domain gap. In addition, we introduce an airway structure awareness loss to enforce depth consistency within the airway lumen while preserving smooth transitions and structural integrity. By incorporating anatomical priors, Brea-Depth enhances model generalization and yields more robust, accurate 3D airway reconstructions. To assess anatomical realism, we introduce Airway Depth Structure Evaluation, a new metric for structural consistency. We validate BREA-Depth on a collected ex-vivo human lung dataset and an open bronchoscopic dataset, where it outperforms existing methods in anatomical depth preservation
An investigation of conservation of fluid oscillation in a continuous oscillatory baffled reactor
No full textContinuous oscillatory baffled reactors (COBRs) have been utilised in organic synthesis and crystallisation, however, no validation work has yet been conducted to determine if the fluid displacement caused by the oscillation at the start of the COBR is conserved at the end of the COBR. This work reports, for the first time, both experimental measurements and theoretical evaluations of the displacement. The experimental validation involves physically measuring the fluid displacement using a laser distance sensor located at the end of the COBR. Surprisingly, the measured displacement values at some operating conditions differ from the initial settings. The theoretical evaluation entails the determination of power dissipation across the COBR using the pressure measurements at four locations along the COBR. The model evaluated displacements agree well with the experimental measurements at all operational conditions, validating the methodologies used in this work
Beyond Predation: Potential Metabolic Roles of Intracellular Bacteria in Acanthamoeba Ecology:Beyond the Trojan Horse
No full textAlthough Acanthamoeba is well known as a reservoir and "Trojan horse" for other microbes, its relationship with intracellular organisms may extend beyond protection. Here, we discuss that certain bacteria contribute metabolically to the host, breaking down complex substrates and providing nutrients that expand its ecological adaptability. The proposed model reframes amoebae not only as predators and shelters, but also as metabolic consortia, with implications for environmental microbiology, protist ecology, and the evolution of opportunistic pathogens. Further studies using integrated multi-omics and co-culture approaches, combining metagenomic and metabolomic profiling of Acanthamoeba-bacteria interactions and transcriptomic analyses will help identify bidirectional metabolic exchange and functional gene expression within the symbiosis.</p
Parametric estimation of conditional Archimedean copula generators for censored data
No full textA novel framework is introduced for estimating Archimedean copula generators in a conditional setting by embedding endogenous variables directly within the generator function. Unlike standard copula constructions that rely on a fixed dependence structure across all covariate levels, the proposed methodology allows both the strength and the shape of dependence to evolve with the covariates. To identify the values of a continuous risk factor at which the dependence pattern undergoes substantive changes, an iterative splitting algorithm is developed to determine optimal partitioning points within the covariate range. The approach is evaluated through applications to a diabetic retinopathy study and a claims reserving analysis, illustrating that explicitly modelling covariate effects yields a more accurate representation of dependence and enhances the practical relevance of copula models in medical and actuarial settings
Access Repository Dashboard
Do you manage Heriot Watt Pure? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!